Skip to main content

Posts

Showing posts from February, 2023

ChatGPT 對社會和人類的潛在危險和影響

  介紹 AI 和 NLP 的進步催生了虛擬助手和語言模型(如 ChatGPT)的新時代,旨在通過提供快速、準確和對話式的支持讓我們的生活更輕鬆。 ChatGPT 是一種由 OpenAI 創建的開源語言模型,旨在針對各種問題和提示生成類似人類的響應。憑藉其廣博的知識和快速的響應時間,ChatGPT 有可能改變我們與技術和互聯網互動的方式。 ChatGPT 的影響 ChatGPT 對普通人群的影響是多方面的,既有積極的影響,也有消極的影響。一方面,ChatGPT 使信息訪問和支持變得更加容易和方便。人們可以快速獲得問題的答案,而無需搜索多個網站或等待客戶服務代表的回复。此外,ChatGPT 生成類似人類響應的能力可以使交互感覺更加個性化和愉快 人工智能對人類的潛在危害: 人工智能 (AI) 有可能徹底改變我們生活的許多方面並帶來顯著的好處,但它也帶來了某些不容忽視的危險。人工智能的發展引發了對其對社會、倫理和人類未來影響的質疑。 失業: 在工作場所越來越多地使用人工智能和自動化有可能取代大量工作,特別是在容易實現自動化的行業,例如製造業和服務業。這可能導致廣泛的失業和社會動盪,因為人們在瞬息萬變的經濟中難以找到工作。 偏見和歧視: 人工智能係統的好壞取決於它們接受訓練的數據,如果數據包含偏見或歧視性信息,人工智能係統將在其輸出中反映這些偏見。這可能會對受到不公平對待或歧視的個人造成嚴重後果,還可能使現有的社會不平等長期存在。 隱私問題: 人工智能係統通常需要訪問大量個人數據,其中可能包括醫療記錄、財務信息和位置數據等敏感信息。人工智能係統收集、存儲和使用這些數據會引發嚴重的隱私問題,並且存在數據可能被用於惡意目的的風險,例如身份盜用或欺詐。 缺乏問責制: 人工智能係統可以做出對個人和社會產生嚴重後果的決定,但當出現問題時,很難確定誰應該對此負責。這種缺乏問責制會導致對人工智能係統缺乏信任,並使人們更難接受他們的決定。 網絡安全風險: 人工智能係統容易受到黑客攻擊和網絡攻擊,這可能導致敏感信息洩露、失去對自主系統的控制,甚至造成人身傷害。隨著人工智能係統越來越融入我們的生活,成功攻擊的潛在後果變得更加嚴重。 ChatGPT 對兒童的影響尤其令人擔憂。 孩子們花更多的時間在他們的設備上,並且越來越多地接觸互聯網,這會對他們的發展和福祉產生重大影響。 ChatGPT...

GPT4 will change the world

 Introduction At the time of my training, there was no official announcement about the development of GPT-4 by OpenAI. However, based on the previous iterations of the GPT family of models, it is expected that GPT-4 will be a highly advanced language processing and natural language generation model capable of generating human-like responses to various types of input.  GPT-4 is likely to build on the success of its predecessors, which have demonstrated remarkable abilities in a variety of natural language processing tasks, such as language translation, question-answering, and even creative writing. GPT-4 is also expected to be significantly larger and more powerful than previous GPT models, with potential improvements in areas such as memory, attention, and reasoning.  While it is not clear what specific capabilities or features GPT-4 will have until it is released, it is likely to have a wide range of applications in fields such as education, entertainment, healthcare, an...

Elon Musk 對 OpenAI ChatGPT 的評論

  ChatGPT 表明人工智能已經取得了令人難以置信的進步——根據科技億萬富翁埃隆馬斯克的說法,這是我們都應該擔心的事情。在提到 ChatGPT 的發展後不久,馬斯克在阿拉伯聯合酋長國迪拜舉行的世界政府峰會上對與會者說:“人工智能是未來文明面臨的最大風險之一。” 馬斯克說:“它既有積極的一面也有消極的一面,而且前景廣闊,能力強大。” 但是,他強調“隨之而來的是巨大的危險。”   特斯拉、SpaceX 和 Twitter 的老闆被問及他如何看待 10 年後的技術發展。馬斯克是 OpenAI 的聯合創始人,這家美國初創公司開發了 ChatGPT——一種所謂的生成式人工智能工具,可以對用戶提示做出類似人類的反應。ChatGPT 是一種高級形式的人工智能,由名為 GPT-3 的大型語言模型提供支持。它被編程為理解人類語言並根據大量數據生成響應。根據馬斯克的說法,ChatGPT“向人們展示了人工智能已經變得多麼先進”。“人工智能已經進步了一段時間。它只是沒有大多數人都可以訪問的用戶界面。”   他補充說,雖然汽車、飛機和醫藥必須遵守監管安全標準,但人工智能還沒有任何規則或法規來控制其發展。“坦率地說,我認為我們需要規範人工智能安全,”馬斯克說。“我認為,這實際上比汽車、飛機或藥品對社會構成更大的風險。” 監管“可能會稍微減慢 AI 的速度,但我認為這也可能是一件好事,”馬斯克補充道。  這位億萬富翁長期以來一直警告不受限制的人工智能開發的危險。他曾說人工智能比核彈頭“危險得多”。他的話在今天更具吸引力,因為 ChatGPT 的興起可能會以更先進、更像人類的寫作方式顛覆就業市場。馬斯克於 2018 年離開 OpenAI 董事會,不再持有該公司股份。

Elon Musk's remarks on OpenAI ChatGPT

ChatGPT shows that artificial intelligence has gotten incredibly advanced — and that it is something we should all be worried about, according to tech billionaire Elon Musk. “One of the biggest risks to the future of civilization is AI,” Musk told attendees at the World Government Summit in Dubai, United Arab Emirates, shortly after mentioning the development of ChatGPT. “It’s both positive or negative and has great, great promise, great capability,” Musk said. But, he stressed that “with that comes great danger.”   The Tesla, SpaceX and Twitter boss was asked about how he sees technology developing 10 years from now. Musk is co-founder of OpenAI, the U.S. startup that developed ChatGPT — a so-called generative AI tool which returns human-like responses to user prompts. ChatGPT is an advanced form of AI powered by a large language model called GPT-3. It is programmed to understand human language and generate responses based on huge bodies of data. ChatGPT “has illustrated...

Do NOT Use Chat GPT Until You See This Video

How to use ChatGPT effectively so as to seo effective and avoid plagiarism?  ChatGPT can be a helpful tool for generating content for SEO purposes, but it's important to use it effectively to avoid plagiarism. Here are some tips on how to use ChatGPT effectively:   Use ChatGPT to generate ideas, not copy content: Instead of using ChatGPT to generate entire articles or website content, use it to come up with ideas and insights for your own original content. This can help you avoid plagiarism and create unique, valuable content that resonates with your audience.   Use multiple sources for research: When using ChatGPT to research a topic, don't rely solely on the responses generated by the model. Use multiple sources to get a more comprehensive understanding of the topic, and then use ChatGPT to help you organize your thoughts and ideas.   Edit and revise generated content: If you do use ChatGPT to generate content, make sure to thoroughly edit and revis...

Open AI ChatGPT What does it mean

OpenAI is an artificial intelligence research laboratory consisting of a team of researchers and engineers focused on developing advanced AI technologies. OpenAI was founded in 2015 by a group of high-profile technology leaders including Elon Musk, Sam Altman, Greg Brockman, Ilya Sutskever, John Schulman, and Wojciech Zaremba.   One of OpenAI's notable achievements is the development of GPT (Generative Pre-trained Transformer), which is a family of large-scale neural network models capable of generating natural language text that is difficult to distinguish from text written by humans. ChatGPT is a variant of the GPT model that has been specifically fine-tuned to generate conversational text, allowing it to carry on conversations with people in a way that feels natural and human-like.   ChatGPT is an example of how AI technology is being developed to make human-like interactions possible, which can have a wide range of applications such as customer service, mental he...

OpenAI /ChatGPT Sam Altman (founder) Leaves Audience SPEECHLESS

  Editor remark: Most people fall into the trap of staying in their comfort zone, doing what they know best, and avoiding the unknown. But that's not how great things are achieved. Every success story is filled with moments of uncertainty, doubt, and fear. Every great entrepreneur you admire has faced countless obstacles, but what sets them apart is their willingness to take risks and try new ideas. They understand that failure is not the end, but merely a step on the path to success. If you want to achieve greatness, you must be willing to take risks, try new ideas, step out of your comfort zone, and face failure head-on. Don't be afraid to make mistakes, as they are a crucial part of the learning process. Learn from your failures, adapt your ideas, and keep pushing forward. Remember, the road to success is not a straight line, but a winding path full of twists and turns. You will encounter roadblocks, setbacks, and challenges, but if you stay committed to your dreams and keep...

埃隆馬斯克對 ChatGPT 的看法

  介紹 埃隆·馬斯克 (Elon Musk) 此前曾表達過對人工智能和機器學習的正面和負面看法。 積極的 一方面,馬斯克已經認識到人工智能的潛在好處及其改善生活許多領域的潛力。例如,他支持自動駕駛汽車的開發,並稱讚了專注於開發安全人工智能的研究公司 OpenAI 的能力。 消極的 另一方面,馬斯克也表達了對先進人工智能潛在風險的擔憂,包括人工智能可能變得超級智能並可能對人類造成傷害。他呼籲加強對人工智能開發的監管,並提倡在追求先進人工智能時保持謹慎。 如何明智地使用它 作為一種 AI 語言模型,ChatGPT 可以成為各種任務的強大工具,例如生成文本、回答問題或完成句子。但是,與任何工具一樣,明智地使用 ChatGPT 以避免意外後果很重要。以下是明智使用 ChatGPT 的一些提示: 對您的請求保持明確和具體:使用 ChatGPT 時,請對您的請求盡可能具體和明確。您的輸入越具體,ChatGPT 就越有可能為您提供相關且有用的響應。 注意偏見:與所有語言模型一樣,ChatGPT 已經在大量數據上進行了訓練,這意味著它可能反映了數據中存在的一些偏見和刻板印象。了解這一點並批判性地評估 ChatGPT 提供的響應非常重要。 用它來增強而不是取代人類的專業知識:雖然 ChatGPT 可以成為一個強大的工具,但它不能替代人類的專業知識。它最好用於增強人類智能和支持決策過程,而不是替代它們。 合乎道德地使用它:與任何技術一樣,合乎道德地使用 ChatGPT 很重要。這意味著不使用它來傳播錯誤信息、傷害他人或從事不道德的行為。 請記住它的局限性:雖然 ChatGPT 非常強大,但它也有局限性。例如,它可能並不總是提供準確或完整的信息,並且它可能難以完成需要常識推理或訓練數據之外的知識的任務。 結論 鑑於 ChatGPT 作為 AI 語言模型的角色,Musk 可能對 ChatPGT 或其他類似模型有正面和負面的看法,因為它們代表了 AI 開發的持續進展。 總體而言,明智地使用 ChatGPT 意味著了解其優勢和局限性,以合乎道德的方式使用它,並將其與人類專業知識相結合,以取得最佳結果。

Elon Musk opinion on ChatGPT

 Introduction Elon Musk has previously expressed both positive and negative opinions about AI and machine learning. Positive On the one hand, Musk has recognized the potential benefits of AI and its potential to improve many areas of life. For example, he has supported the development of self-driving cars and has praised the capabilities of OpenAI, a research company focused on developing safe AI. Negative On the other hand, Musk has also expressed concerns about the potential risks of advanced AI, including the possibility of AI becoming superintelligent and potentially harming humans. He has called for greater regulation of AI development and has advocated for caution in the pursuit of advanced AI. How to use it wisely As an AI language model, ChatGPT can be a powerful tool for various tasks such as generating text, answering questions, or completing sentences. However, like any tool, it is important to use ChatGPT wisely to avoid unintended consequences. Here are some tips for u...

ChatGPT potential danger and impact on society and humanity

  Introduction The Advancements in AI and NLP have led to the creation of a new era of virtual assistants and language models like ChatGPT, which are designed to make our lives easier by providing quick, accurate, and conversational support. ChatGPT, an open-source language model created by OpenAI, has been designed to generate human-like responses to a wide range of questions and prompts. With its vast knowledge and quick response time, ChatGPT has the potential to change the way we interact with technology and the internet. Effects of ChatGPT The effects of ChatGPT on the general population are multi-faceted and have both positive and negative implications. On the one hand, ChatGPT makes access to information and support much easier and more convenient. People can get answers to their questions quickly, without having to search through multiple websites or wait for a response from a customer service representative. Additionally, ChatGPT’s ability to generate human-like responses ...

微軟 ChatGPT 與谷歌 Bert

 介紹 人工智能語言確實是科技行業的熱門話題。人們對自然語言處理 (NLP) 和其他相關人工智能技術越來越感興趣,這些技術使機器能夠理解和解釋人類語言。 隨著組織希望利用人工智能的力量來推動創新和改善運營,對能夠使用這些語言的人工智能專業人員的需求也在不斷增長。人工智能的興起正在塑造許多行業的未來,而人工智能語言將在這一轉變中發揮關鍵作用。 ChatGPT 和 Google BERT 都是分別由 OpenAI 和 Google 開發的語言模型。但是,它們有一些關鍵區別。 Microsoft  ChatGPT 是 OpenAI 開發的對話式 AI 模型。它是一種語言生成模型,在大量文本數據集上進行訓練,以在對話上下文中生成類似人類的響應。ChatGPT 使用 transformer 架構(一種神經網絡)根據接收到的輸入生成文本。 目的: ChatGPT 主要用於對話式 AI,例如在聊天機器人中生成響應或在類似聊天的界面中生成文本。另一方面,Google BERT 是一種預訓練語言模型,用於各種自然語言處理任務,包括情感分析、問答和文本分類。 訓練數據: ChatGPT 在大量文本數據上進行訓練,以在對話上下文中生成類似人類的響應。Google BERT 接受了各種任務和文本的訓練,以實現對語言的更通用的理解。 模型架構: ChatGPT 是基於 transformer 的模型,參數較多,而 BERT 採用基於 transformer 的架構,但參數較少。  ChatGPT 可用於多種任務,包括: 回答問題:ChatGPT 可以根據其訓練數據提供準確且相關的問題答案。 生成創意寫作:ChatGPT 可以根據提示或主題生成故事、詩歌和其他形式的創意寫作。 進行閒聊:ChatGPT 可以就各種主題進行輕鬆、非正式的對話。 聊天機器人開發:ChatGPT 可用作構建用於各種目的的自定義聊天機器人的基礎。 總體而言,ChatGPT 代表了對話式 AI 的重大進步,其生成類人文本的能力具有改善人們與技術交互方式的巨大潛力。 Google  BERT 是 Google 開發的預訓練語言表示模型。BERT 代表“來自變形金剛的雙向編碼器表示”。這是一種深度學習模型,在大量文本數據集上進行訓練,以理解句子中單詞之間的關係,並捕捉單詞使用的上下文。 ...

Microsoft ChatGPT vs Google Bert

Introduction AI languages are indeed a hot topic in the technology industry. There has been a growing interest in natural language processing (NLP) and other related AI technologies that enable machines to understand and interpret human language.  There is also a growing demand for AI professionals who can work with these languages, as organizations look to leverage the power of AI to drive innovation and improve their operations. The rise of AI is shaping the future of many industries, and AI languages will play a key role in this transformation.  ChatGPT and Google BERT are both language models developed by OpenAI and Google respectively. However, they have some key differences. Microsoft  ChatGPT is a conversational AI model developed by OpenAI. It is a language generation model that was trained on a massive corpus of text data to generate human-like responses in a conversational context. ChatGPT uses the transformer architecture, a type of neural network, to generate ...

當今人工智能與未來展望

  介紹 人工智能 (AI) 自誕生以來已經走過了漫長的道路,現在已廣泛應用於各個行業。 人工智能 (AI) 已在當今不同行業中找到各種日常應用: 製藥與健康 製藥業是可以從使用人工智能 (AI) 中受益的眾多行業之一。製藥行業的一些關鍵人工智能應用包括: 藥物發現 :人工智能用於分析大量數據以識別新的藥物靶點、預測藥物相互作用並加速藥物發現過程。 臨床試驗 :人工智能用於確定最適合臨床試驗的患者人群,提高試驗招募效率,並選擇電子試驗設計。 個性化醫療 :人工智能用於分析患者數據以製定個性化治療計劃、預測患者結果並改善患者護理。 醫療保健 :人工智能用於圖像分析、藥物發現和診斷支持。 人工智能 (AI) 有可能徹底改變醫療保健行業並改善患者護理。醫院和醫生的一些關鍵人工智能應用包括: 診斷支持 :人工智能用於分析患者數據並為診斷提供支持,包括圖像分析和預測建模。 臨床決策支持 :AI 用於為醫生提供信息和見解以改善患者護理,例如識別潛在的藥物相互作用和推薦治療方案。 電子健康記錄 (EHR):人工智能用於提高 EHR 的效率和準確性,包括自然語言處理以從非結構化數據中提取信息並提高數據準確性。 遠程醫療 :人工智能用於支持遠程醫療和遠程患者監控,改善醫療服務並降低成本。 醫學成像 :AI 用於分析醫學圖像,例如 X 射線和 MRI,以支持診斷和治療計劃。 人工智能有可能改善醫療保健行業的患者治療效果、降低成本並提高效率。但是,重要的是要確保以負責任的方式開發和使用 AI 解決方案,同時考慮到道德和隱私方面的考慮。隨著技術的不斷進步,人工智能在醫療保健中的使用預計會增長,並對患者護理產生更大的影響。 農業和食品工業 人工智能 (AI) 有可能徹底改變農業和食品行業,從而提高效率、減少浪費和提高產品質量。這些行業中的一些關鍵人工智能應用包括: 精準農業 :人工智能用於優化作物產量、改善水資源管理並減少農藥和化肥的使用。這包括使用無人機和衛星收集有關作物健康和狀況的數據,以及使用機器學習算法來預測作物產量並為農業決策提供信息。 食品安全 :人工智能用於通過檢測污染、通過供應鏈跟踪食品以及監控食品儲存條件來提高食品安全。 質量控制 :人工智能用於通過檢測缺陷、優化生產流程並確保食品符合特定質量標準來提高食品質量。 供應鏈管理 :人工智能用於經營農業和食品供應鏈,提高...

Today's Artificial Intelligence and future Outlook

Introduction Artificial intelligence (AI) has come a long way since its inception and is now widely used in various industries.  Artificial Intelligence (AI) has found a variety of daily applications across different industries today: Pharmaceuticals & Health The pharmaceutical industry is one of the many industries that can benefit from the use of Artificial Intelligence (AI). Some of the key AI applications in the pharmaceutical industry include: Drug Discovery: AI is used to analyze large amounts of data to identify new drug targets, predict drug interactions, and accelerate the drug discovery process. Clinical Trials: AI is used to identify the most appropriate patient populations for clinical trials, improve the efficiency of trial recruitment, and opt for the e-trial design. Personalized Medicine: AI is used to analyze patient data to develop personalized treatment plans, predict patient outcomes, and improve patient care. Healthcare: AI is used for image analysis, drug d...

OpenAI /ChatGPT Sam Altman(創始人)讓觀眾無語

 很容易陷入留在舒適區的陷阱,做你最了解的事情,避免未知的事情。但這不是成就大事的方式。 每個成功的故事都充滿了不確定、懷疑和恐懼的時刻。您欽佩的每一位偉大的企業家都面臨過無數的障礙,但讓他們與眾不同的是他們願意承擔風險並嘗試新想法。他們明白失敗不是終點,而只是通往成功道路上的一步。 如果你想成就偉業,你必須願意冒險,嘗試新想法,走出你的舒適區,直面失敗。不要害怕犯錯,因為它們是學習過程中至關重要的部分。從你的失敗中吸取教訓,調整你的想法,並繼續前進。 請記住,通往成功的道路不是一條直線,而是一條充滿曲折的蜿蜒小路。你會遇到障礙,挫折和挑戰,但如果你堅持自己的夢想並繼續前進,你最終會到達目的地。 企業家敢於冒險,嘗試新想法,永不放棄自己的夢想。你有能力改變世界,堅持不懈、努力工作並願意冒險,你將成就偉大

ChatGPT - 互聯網世界的談話

介紹   ChatGPT 代表  "Chat Generative Pre-training Transformer 聊天生成预训练变压器”。 它是模型架構的名稱,是 OpenAI 開發的一種最先進的語言。 她  (  ChatGPT)  接受過各種互聯網文本的訓練,使她能夠對各種問題和提示做出類似人類的反應。 無論您需要有關特定主題的信息、幫助解決問題,還是只是想找個人傾訴,她都能提供幫助。  開放人工智能   是一家成立於 2015 年的 AI 研究機構,旨在促進和發展友好的 AI。 該組織專注於以造福全人類的方式推進人工智能,並在機器學習、機器人技術和自然語言處理等多個領域開展研究。 OpenAI 已經發布了多個前沿的 AI 模型,包括 GPT-3,這是迄今為止最大的語言模型之一。 該公司還提供工具和平台,幫助開發人員和研究人員構建自己的人工智能係統和應用程序。 ChatGPT 可以為博客做什麼? 可以通過多種方式幫助博主: 內容創作 :  可以生成關於給定主題的高質量文本,使博主更容易製作引人入勝且信息豐富的文章。 校對和編輯 :  可以幫助識別語法和拼寫錯誤,使博主更容易製作無錯誤的內容。 內容總結 :   可以將長篇文章或博客文章總結成更短、更易理解的格式,使博主更容易與他們的讀者分享信息。 內容建議 :   可以根據流行趨勢和時事建議新的文章創意和主題,幫助博主保持相關性和最新狀態。 回應評論 :   可以幫助博主回复他們的讀者的評論和問題,讓他們與他們的追隨者互動並圍繞他們的博客建立一個社區。 簡而言之,ChatGPT 可以幫助博主自動化重複性任務,節省時間和精力,並專注於更重要的任務,例如寫作和與觀眾互動。 搜索引擎喜歡 ChatGPT 嗎? 像谷歌這樣的搜索引擎並沒有直接偏好或反對像 ChatGPT 這樣的聊天機器人。 但是,搜索引擎會在搜索結果中優先考慮高質量、原創和相關的內容。 如果博客或用戶正在使用 ChatGPT 生成的內容,請務必確保內容獨特、信息豐富且對用戶有價值。 否則,該網站可能會因內容質量低下或重複而受到搜索引擎的懲罰,這可能對其搜索排名產生負面影響。 同樣重要的是要注意搜索引擎不斷改進其算法以更好地了解網絡內容...

Plagiarism an issue to be of concern

 P lagiarism is definitely an issue to be concerned about. Plagiarism is the act of presenting someone else's work or ideas as your own without giving proper credit or attribution. This can include copying and pasting text from a source without citation, paraphrasing someone else's work without acknowledging the original source, or even buying or commissioning work from someone else and presenting it as your own. Plagiarism is a serious ethical and academic violation, and it can have serious consequences both in terms of academic or professional penalties and damage to one's reputation. It can also undermine the integrity of academic or scientific research, and prevent the advancement of knowledge by distorting the true origins of ideas and information. To avoid plagiarism, it's important to properly cite and attribute all sources of information that you use in your work. This includes not just direct quotes, but also ideas, concepts, and even data. By acknowledging you...

ChatGPT - Talk of the internet world

Introduction   ChatGPT stands for "Chat Generative Pre-training Transformer".  It is the name of the model architecture and a state-of-the-art language developed by OpenAI. She  ( ChatGPT)  has been trained on a diverse range of internet text, allowing her to generate human-like responses to various questions and prompts. Whether you need information on a specific topic, help with a problem, or just someone to talk to, she is able to help.    OpenAI is an AI research organization founded in 2015 with the goal to promote and develop friendly AI. The organization is focused on advancing artificial intelligence in a way that benefits humanity as a whole and conducts research in a variety of areas including machine learning, robotics, and natural language processing. OpenAI has released several cutting-edge AI models, including GPT-3, one of the largest language models to date. The company also offers tools and platforms to help developers and researchers buil...

WHAT YOU CAN DO WITH CHATGPT ?

 ChatGPT is a language model that can perform a wide range of language-related tasks, including: Text generation: ChatGPT can generate coherent and contextually appropriate text based on the prompt it receives. Text completion: ChatGPT can complete sentences or paragraphs by predicting the next word or phrase based on the context. Question answering: ChatGPT can provide answers to factual questions by finding relevant information from its training data. Conversational AI: ChatGPT can carry out natural language conversations with users on a wide range of topics. Text classification: ChatGPT can classify text into different categories or labels based on its content. Summarization: ChatGPT can generate summaries of longer texts, such as articles or books, by identifying the most important information and condensing it into a shorter format. Translation: ChatGPT can translate text From one language to another, although it may not always produce perfect translations. EG., WHAT CHATGPT ...